[Gluster-devel] Verify infinite loop bug with ext4

Christian Wittwer wittwerch at gmail.com
Tue Oct 9 10:00:29 UTC 2012


We have currently a strange behaviour on our gluster 3.3 volume (4
bricks, distributed+replicated). From time to time the cpu usage
increases to around 50% and the gluster daemon starts responding.
After some research I got a pointer to a bug report
(https://bugzilla.redhat.com/show_bug.cgi?id=838784). We use kernel
3.2.0-26-generic on Ubuntu 12.04, so I'm afraid we are affected.
I used the test script, but I'm not able to interpret the results correct.

root at foo:~# ./test /data/brick0/instances/_base/
496e4306f1de9b647fc0a6021b35fd5428e88c70_10: 495888329 (495888329)
..: 616848399 (616848399)
.: 723397576 (723397576)
fc9ab12b1551f9c358b5acd65e8fcbe197c89f6c: 752979281 (752979281)
ephemeral_0_40_None: 995810121 (995810121)
496e4306f1de9b647fc0a6021b35fd5428e88c70: 2147483647 (2147483647)

Does that mean I'm affected by the ext4 bug?
If not, what can I do to gather enough information to open a new bug report?

Regards,
Chris




More information about the Gluster-devel mailing list